RFR: 8317763: Follow-up to AVX512 intrinsics for Arrays.sort() PR [v5]

Vladimir Kozlov kvn at openjdk.org
Thu Oct 12 00:38:21 UTC 2023


On Wed, 11 Oct 2023 23:40:55 GMT, Sandhya Viswanathan <sviswanathan at openjdk.org> wrote:

> > It makes sense to let `-XX:ControlIntrinsic=` overrule `VM_Version::is_intel()` check and enable the intrinsics when `AVX512DQ` is supported.
> 
> May be it could be done as part of https://bugs.openjdk.org/browse/JDK-8317976.

It is not easy to do. You need to make sure that you not enable intrinsic by `-XX:ControlIntrinsic=` when it is not supported by hardware. I would say it is separate (second) RFE because we need to test it separately from current Zen 4 issue.

Note, we are not introducing regression to AMD with these changes - code is reverted back to before previous changes [14227](https://github.com/openjdk/jdk/pull/14227). We have some  time to propose patch.

-------------

PR Comment: https://git.openjdk.org/jdk/pull/16124#issuecomment-1758737716


More information about the build-dev mailing list